Surgical Outcomes of 25-Gauge Pars Plana Vitrectomy for Diabetic Tractional Retinal Detachment

1 1 Surgical outcomes of MS Dikopf , KH Patel , VJ Setlur and JI Lim STUDY CLINICAL 25-gauge pars plana vitrectomy for diabetic tractional retinal detachment

Abstract Purpose To evaluate the outcomes and Introduction complications of 25-gauge (G) pars plana Tractional retinal detachment (TRD) is a major vitrectomy (PPV) for repair of diabetic cause of vision loss in patients with proliferative tractional retinal detachment (TRD). diabetic retinopathy (PDR).1,2 In about 5–10% Methods Retrospective review of of cases of retinal detachment, anatomical consecutive, single-surgeon 25-G PPV cases and visual prognoses are complicated by between July 2007 and July 2014. Seventy proliferative vitreoretinopathy (PVR).3,4 Surgical eyes from 55 patients were operated on for management of diabetic TRD, with or without diabetic TRD; all eyes were tamponaded with sulfur hexafluoride, octofluoropropane, PVR, may often be challenging. silicone oil, or balanced salt solution. Mean Repair of complicated vitreoretinal disease, Retina Service, UIC Department of age at surgery was 47.7 years (range 23–76 such as diabetic TRD, has conventionally been accomplished via 20-gauge (G) or 23-G pars Ophthalmology and Visual years), and mean length of follow-up was Sciences, Illinois Eye and Ear 713 days (range 90–2368 days; median plana vitrectomy (PPV) systems. However, the Infirmary, University of fi 671 days). Primary outcomes included best- 25-G pars plana vitrectomy (PPV) system, rst Illinois Hospital & Health 5 corrected visual acuity (BCVA), intraocular described by Fujii and colleagues in 2002, offers Sciences System, Chicago, IL, USA pressure (IOP), anatomic success, many advantages over 20- or 23-G PPV. These redetachment, and endophthalmitis. include reduced operative times, reduced post- fl Correspondence: Results Preoperatively, 49 eyes (70%) had a operative in ammation, faster visual recovery, JI Lim, Retina Service, UIC concurrent rhegmatogenous component (8 of increased patient comfort, reduced rate of Department of which also had proliferative vitreoretinopathy intraoperative retinal break formation, Ophthalmology and Visual (PVR)). Mean BCVA improved from avoidance of conjunctival dissection and limbal Sciences, Illinois Eye and Ear Infirmary, University of stem cell damage, and less induced corneal logarithm of the minimal angle of resolution Illinois Hospital & Health 5–12 1.59 (20/800, SD 0.88) to 0.68 postoperatively astigmatism. Sciences System, 1855 W. (20/100, 0.77), P-valueo0.001. Mean IOP Upon its advent, 25-G PPV was applied Taylor Street, Suite 2.50, increased from 15.9 to 20 mm Hg 1 day after towards treating macular holes, idiopathic Mail Code 648, Chicago, IL surgery. Elevated postoperative IOP epiretinal membranes, refractory macular 60611, USA Tel: +1 312 413 0704; ≥ edema, and non-clearing vitreous hemorrhage.13 ( 22 mm Hg) occurred in 25 eyes, and low Fax +1 312 413 7929. IOP (≤ 5 mm Hg) occurred in 2 eyes. Primary There were concerns that fragility or inadequate E-mail: [email protected] reattachment was achieved in 63 eyes (90%), variety of instrumentation may limit its and final anatomical success occurred in 69 usefulness in more complicated surgical 1These authors contributed eyes (99%). There were no cases of disease.14,15 However, instrumentation has equally to this work. endophthalmitis. expanded,16 and stiffer 25-G instrumentation has Received: 6 January 2015 Conclusions Twenty-five-G PPV repair led to successful outcomes of 25-G PPV for a Accepted in revised form: 6 was safe and effective in the repair of wide variety of vitreoretinal pathologies. 26 May 2015 diabetic TRD, including eyes with a Given the benefits of 25-G systems, several Published online: combined rhegmatogenous detachment or studies have investigated its use in the treatment 17 July 2015 PVR. Gas, silicone oil, and balanced salt of complicated retinal detachment, including solution tamponading agents all proved to diabetic TRD; comparable outcomes and Preliminary data from this be efficacious in this surgical population. series was presented at the complication rates between 25-G PPV and 20-G 2013 Annual Meeting of the – 6,7,17–20 Eye (2015) 29, 1213 1219; doi:10.1038/eye.2015.126; or 23-G systems have been found. Association for Research in published online 17 July 2015 However, these studies included limited Vision and Ophthalmology. 25-G pars plana vitrectomy for diabetic TRD MS Dikopf et al 1214

numbers of TRD eyes and only evaluated silicone oil or intraocular pressure (IOP) measured by Goldmann octofluoropropane (C3F8) tamponades. This study aims applanation tonometry or a Tono-pen XL applanation to evaluate the effectiveness of 25-G PPV and a variety of tonometer (Reichert Technologies, Depew, NY, USA), tamponade agents in a large population of eyes with and rates of postoperative complications including diabetic TRD, many of which also had combined redetachment, vitreous hemorrhage, epiretinal membrane rhegmatogenous retinal detachments (RRD) or PVR. formation, ocular hypertension or hypotony, and endophthalmitis. Materials and methods BCVA was measured using the standard Snellen visual acuity chart and converted to the logarithm of the We conducted a retrospective review of the medical minimal angle of resolution (LogMAR) visual acuity.21 records of all surgical cases performed by a single Peak BCVA at any time during the postoperative period vitreoretinal surgeon (JIL) at the University of Illinois was used to represent the postoperative LogMAR visual Hospital, between July 2007 and July 2014. The study was acuity. Preoperative and postoperative LogMAR values conducted in compliance with the Health Insurance were compared with a paired, two-tailed t-test, and a Portability and Accountability Act and was approved by P-value o0.05 was considered statistically significant. the Institutional Review Board at the University of Outcomes for all eyes, as well as per form of Illinois, Chicago. Eyes that underwent 25-G PPV for tamponade were performed. Given variability in location primary repair of a diabetic TRD were eligible for and type of retinal detachments, subgroup analysis was inclusion; patients with o3 months of postoperative also performed in regards to macular involvement, follow-up, or patients with other proliferative existence of a rhegmatogenous component, and existence vitreoretinopathy disease (such as proliferative sickle cell of PVR. retinopathy) were excluded. No patients underwent preoperative treatment with anti-VEGF agents. PPV was performed using the Alcon Accurus or Alcon Results Constellation 25-G vitrectomy machines (Alcon Patient demographics Laboratories, Inc., Fort Worth, TX, USA), with a maximum cut rate of 5000 cuts per min and maximum vacuum rate Seventy eyes from 55 patients were included in this study of 650 mm Hg. In cases that required bimanual technique (Table 1). The mean age of all patients was 47.7 years for dissection of extensive fibrovascular membranes, an (range, 23–76 years) and mean length of follow-up was illuminated 25-G pick or a 25-G self-retaining chandelier 713 days (range, 90–2368 days; median 671 days). Of note, illumination fiber was utilized. All cases underwent 6 of our 55 patients were lost to follow-up between 3 and vitrectomy and vitreous base shaving using a wide-angle 6 months after surgery. Four of them were tamponaded viewing system and scleral depression during the vitreous with SF6, 1 with C3F8 and 1 had BSS; none had PVR. base dissection. Perfluro-n-octane (PFO) liquid was used Mean follow-up excluding these patients was 775 days as needed. Endolaser photocoagulation was applied for (range 204–2368 days; median 730 days). All 55 patients treatment of neovascularization or to barricade retinal were included for data analysis. breaks. Fluid-air exchange followed by injection of sulfur fl hexa uoride (SF6), C3F8, silicone oil, or balanced salt solution (BSS) was performed to tamponade retinal breaks Table 1 Patient characteristics or detachments. The choice of tamponade was determined Number of eyes 70 by the extent of detachment, amount of traction, existence Number of patients 55 of a rhegmatogenous component and, in rare cases, the Male/female 23/32 (42%/58%) patient’s preoperative preference for silicone oil instead Mean age at surgery (years) 47.7 Mean length of follow-up (days) 713 of gas. fi Median lenth of follow-up (days) 671 In patients with a preoperative visually signi cant Mean preoperative IOP (mm Hg) 15.9 cataract, cataract extraction and implantation of an Mean postoperative day 1 IOP (mm Hg) 20 acrylic, foldable intraocular lens was performed via pars Eyes phakic preoperatively 66 (94%) plana lensectomy or phacoemulsification through a clear Eyes with lens extracted in combination 10 (14%) with PPV corneal incision. Eyes with PFO used intraoperatively 15 (21%) All sclertomies in this series were sewn to prevent Eyes with preoperative vitreous 53 (76%) against postoperative hypotony or infection, as hemorrhage present discussed later. Mean operative time (min) 118

Primary outcome measures included postoperative Abbreviations: IOP, intraocular pressure; PFO, perﬂuoro-n-octane; best-correct Snellen visual acuity (BCVA), postoperative PPV, pars plana vitrectomy.

Eye 25-G pars plana vitrectomy for diabetic TRD MS Dikopf et al 1215

Table 2 Visual acuity outcomes and anatomical success per form of tamponade

Form of Number Mean preoperative visual Mean peak postoperative P-value Eyes with primary Eyes with ﬁnal anatomical tamponade of eyes acuity (LogMAR) visual acuity (LogMAR) reattachment achieved success achieved o SF6 Gas 22 1.48 (~20/600) 0.51 (~20/60) 0.001 20 (91%) 21 (96%) C3F8 Gas 22 1.49 (~20/600) 0.95 (~20/175) o0.05 18 (82%) 22 (100%) BSS 21 1.69 (~20/1000) 0.42 (~20/50) o0.001 21 (100%) 21 (100%) Silicone Oil 5 2.15 (~20/3000) 1.53 (~20/700) 0.37 4 (80%) 5 (100%) Overall 70 1.59 (~20/800) 0.68 (~20/100) o0.001 63 (90%) 69 (99%)

Abbreviations: BSS, balanced salt solution; LogMAR, logarithm of the minimum angle of resolution.

Mean preoperative IOP was 15.9 mm Hg (range, preoperative BCVA of 2.15 (20/3000, 0.98, 1.29–3.01), and 7–30 mm Hg); preoperative IOP was unavailable for one improved to a mean of 1.53 (20/700,1.10, 0.57–2.49), eye. Preoperative vitreous hemorrhage was present in 53 P = 0.37 (Table 2). Of note, four eyes, one in each eyes. Four eyes were pseudophakic preoperatively, while tamponade group, had pre- or postoperative visual the remaining 66 eyes were phakic-10 had a visually acuities that were light perception or worse, and therefore significant cataract that required extraction during the could not be converted to LogMAR.21 Three eyes had a primary TRD repair. preoperative visual acuity of light perception; two improved postoperatively to 20/400, whereas the other remained at light perception. A fourth eye worsened from Operative findings hand motion to light perception, due to neovascular Forty-nine out of 70 eyes (70%) had a combined glaucoma and recurrent, dense vitreous hemorrhage. rhegmatogenous and tractional retinal detachment, and 8 As shown in Table 3, 49 eyes with a combination of these eyes (11%) also had PVR. During surgery, five tractional and rhegmatogenous detachment improved eyes had a sclerotomy site enlarged to 20-G. In four eyes from a preoperative mean LogMAR BCVA of 1.54 this was performed secondary to instrumentation; three (20/700, 0.85, 1.30–1.80) to 0.80 (20/125, 0.87, 0.56–1.04), required the use of a 20-G fragmentor for lensectomy, Po0.001. Of the 21 eyes with solely a TRD, Snellen acuity whereas another required the use of a 20-G endolaser improved from a mean of 1.7 (20/1000, 0.95, 1.29–2.11) because the 25-G model was unavailable. For the fifth eye, to 0.42 (20/50, 0.40,0.25–0.54), P o0.001. Thirty-nine a sclerotomy was enlarged to acquire a sample of patients had a detachment involving the macula; neovascularization of the disc for research purposes. For preoperative BCVA was 1.61 (20/800, 0.82, 1.35–1.87) and all eyes with an enlarged sclerotomy, a 20–25-G plug improved to 0.86 (20/145, 0.76, 0.62–1.10), Po 0.001. was utilized to allow for continued use of 25-G Thirty-one patients were operated on prior to macular instrumentation. PFO liquid was used for intraoperative involvement; among these eyes, vision improved from a flattening of detached retina in 15 eyes. All eyes were mean of 1.56 (20/725, 0.96, 1.22–1.90) to 0.47 (20/60, 0.74, – o ultimately tamponaded with SF6 (22 eyes), C3F8 (22), BSS 0.21 0.73), P 0.001. Of the eyes that a combined TRD (21), or silicone oil (5). and RRD, 8 also had pre-existing PVR; in this group, Mean operative time was 118 min (range, 52–270 min); preoperative mean BCVA of 2.3 (20/4000, 0.76, 1.77–3.06) operative time was unavailable for six eyes. improved to 1.88 (20/1500,1.25, 1.01–2.75), P = 0.31. Overall, primary reattachment was achieved in 63 eyes (90%). Six eyes redetached, and 5 were reattached with a Surgical outcomes total of six additional surgeries; final anatomical success For all eyes, mean preoperative BCVA improved from was therefore achieved in 69 eyes (99%). One eye was noted LogMAR 1.59 (20/800, SD 0.88, 95% confidence interval to have a macular sparing TRD at the last visit; vision was (CI) 1.38–1.80) to 0.68 (20/100, 0.77, 0.50–0.86) after hand motion and limited by severe retinal ischemia, and surgery, Po0.001. Postoperative BCVA improvements the patient declined further surgical intervention. were found in all forms of tamponade. Eyes with SF6 gas improved from LogMAR 1.48 (20/600, SD 0.64, CI Postoperative complications 1.17–1.76) to 0.51 (20/60, 0.65, 0.30–0.72 ), Po0.001. Eyes with C3F8 gas improved from 1.49 (20/600, 0.99, Mean post-operative day-1 IOP was 20 mm Hg (range, 1.07–1.90) to 0.95 (20/175, 0.93, 0.56–1.34), Po0.05. Eyes 2–45 mm Hg). Postoperative day-1 IOP was unavailable with BSS tamponade improved from 1.69 (20/1000, 0.96, for five eyes, however, all patients were closely followed 1.28–2.1) to 0.42 (20/50,.40, 0.25–0.59), Po0.001. Eyes that and postoperative week-1 IOP values were available underwent silicone oil tamponade had the worst mean (week-1 values were not substituted into the data

Eye 25-G pars plana vitrectomy for diabetic TRD MS Dikopf et al 1216

analysis). Elevated IOP (≥ 22 mm Hg) on postoperative day 1 occurred in 25 eyes. Speciﬁcally, this occurred in 10 nal ﬁ eyes with SF6 gas (45% of eyes in this group), 9 eyes with C3F8 gas (41%), 5 eyes with BSS (24%), and 1 eye with silicone oil (20%). Two of these SF6 eyes had a combined

Eyes with lensectomy (50% of total combined lensectomies in this group), one of the C3F8 eyes (100%), two of the BSS eyes anatomical success achieved (67%), and one of the silicone oil eyes (50%). Most cases of postoperative ocular hypertension were adequately managed with topical medications, however, few complicated cases occurred in each group. One of the eyes with SF6 was noted to have an IOP of 24 mm Hg preoperatively, which rose to 38 on postoperation day 1; this was initially resistant to topical medications

ent; TRD, tractional retinal detachment. necessitating an anterior chamber tap, and intraocular Eyes with primary reattachment achieved pressures were thereafter maintained in the teens with topical medicine. However, 2 months after surgery, this eye was found to have a pressure of 44 mm Hg secondary to neovascular glaucoma, requiring transcleral diode laser 0.001 34 (87%) 39 (100%) 0.001 29 (94%) 30 (97%) 0.001 42 (86%) 48 (98%) 0.001 21 (100%) 21 (100%) -value o o o o cyclophotocoagulation. Further, one eye with C3F8 was found to have an IOP of 40 mm Hg on postoperative day 4, associated with a hyphema. Another eye in this group )P was found to have consistently elevated pressures in the 30 s secondary to development of neovascular glaucoma, leading to implantation of an Ahmed valve 4 months after surgery. One silicone oil eye with elevated postoperative day-1 pressure was found to have a hemorrhagic choroidal effusion. ≤ Mean peak postoperative visual acuity (LogMAR Two eyes were hypotonous (IOP 5 mm Hg) on postoperative day 1 (1 with SF6, 1 with BSS). For both eyes the hypotony self-resolved during the ﬁrst ) postoperative week, without further sequelae. In eyes with SF6 tamponade, primary reattachment of the retina was achieved in 20 eyes (91%); of the two eyes that re-detached, one was successfully repaired while the other was not re-operated on secondary to patient

Mean preoperative preference and severe retinal ischemia from diabetic

visual acuity (LogMAR retinopathy, as aforementioned. In this group, two eyes (9.1%) developed a cataract requiring surgery, eight eyes (36.3%) developed an ERM (one eye required a membranectomy), and four (18.2%) had recurrent vitreous hemorrhage (no eyes required reoperation for this). These outcomes, including final BCVA for reoperated eyes, are shown in Table 4. In eyes with C3F8 gas, primary reattachment was achieved in 18 eyes (82%); all four eyes that had redetachments were successfully repaired with subsequent surgery (five total surgeries for the four eyes). Eight eyes (36.4%) developed a visually significant cataract necessitating removal, three (13.6%) developed fi Surgical outcomes per type of retinal detachment an ERM (one required surgery), and ve developed recurrent vitreous hemorrhage (one requiring surgery). All 21 eyes with BSS tamponade had primary PVR 8 2.3 (~20/4000) 1.88 (~20/1500) 0.31 5 (63%) 8 (100%) Macular involving 39 1.61 (~20/800) 0.86 (~20/145) Macular threatening 31 1.56 (~20/725) 0.47 (~20/60) Combination TRD/RRD 49 1.54 (~20/700) 0.80 (~20/125) TRD Only 21 1.70 (~20/1000) 0.42 (~20/50) Table 3 Abbreviations: LogMAR, logarithm of the minimum angle of resolution; PVR, proliferative vitreoretinopathy; RRD, rhegmatogenous retinal detachm Type Number of eyes reattachment. Three eyes (14.%) developed cataract with

Eye 25-G pars plana vitrectomy for diabetic TRD MS Dikopf et al 1217

subsequent extraction, five (23.8%) developed an ERM (none needing surgery by the conclusion of this study), and seven (33.3%) had recurrent vitreous hemorrhage (LogMAR) Final BCVA (three requiring reoperation). Of the five eyes with silicone oil tamponade, primary reattachment was achieved in four (80%), and the eye that redetached was successfully repaired with one additional surgery. Two eyes in this group had a combined lensectomy with primary surgery, and an additional eye developed a visually significant cataract requiring surgery. One eye in this group developed an ERM, and one had a recurrent vitreous hemorrhage (neither with further surgery). At their latest visit, four eyes had no light perception vision but were attached anatomically; preoperative visual acuity was hand motion for these eyes, and they had progressive visual loss secondary to retinal ischemia m angle of resolution; VH, vitreous hemorrhage; PPV, pars plana and neovascular glaucoma. There were no cases of postoperative endophthalmitis observed during this study.

Discussion

Use of 25-G PPV is gaining popularity in the treatment of (LogMAR) ERM formation ERM requiring Mx Final BCVA a wide range of surgical vitreoretinal conditions. In regard to complicated TRD repair, 25-G PPV may offer several theoretical benefits over larger gauge systems. Smaller gauge PPV has more stable fluidics from diminished flow (smaller radius of 25-G tubing), which, with high cut rates, may minimize the risk of iatrogenic Recurrent VH requiring PPV 16

percent of total eyes per tamponade subgroup. retinal breaks. Breaks may also be avoided via smaller a microcannulas, as they may better preserve the vitreous base and prevent vitreal incarceration and dragging.10 Twenty-ﬁve-gauge instrumentation also allows easier access to small areas, providing safer vitrectomy and more precise dissection of membranes.12,16 Recurrent VH Still, there remain documented concerns with the use of 25-G PPV for complicated surgical cases, including diabetic TRD. Some worry that the ﬂexibility of 25-G ) instrumentation may prevent adequate rotation of the eye, limiting the removal of peripheral vitreous or

(LogMAR preretinal diabetic membranes and ultimately leading Final BCVA to recurrent detachments.12,16 However, stiffer instrumentation is now available. In our series, redetachment rates were low, as 63 of 70 eyes (90%) achieved primary reattachment and ﬁnal anatomical ) 0.15 (~20/30) 4 (18.2%) 0 8 (36.3%) 1 0.4 (~20/50) a success was achieved in 69 of 70 eyes (99%). Another concern with (25G) sutureless PPV surgery is postoperative hypotony (here deﬁned as IOP requiring CE/IOL Cataract formation ≤ 5 mm Hg).16 In a study by Bamonte et al,22 ~ 9.2% of eyes were hypotonous 1 day after primary 25-G

Postoperative complications per form of tamponade vitreoretinal surgery; higher rates were incidentally found with ﬂuid tamponades.22,23 In our study, two eyes were

gas 2 (9.1% hypotonous on postoperative day 1—one eye with SF6 6 SF vitrectomy; ERM, epiretinal membrane; Mx, membranectomy; BSS, balanced salt solution. Table 4 Tamponade Abbreviations: CE/IOL, cataract extraction/intraocular lens implantation; BCVA, best-corrected visual acuity; LogMAR, logarithm of the minimu C3F8 gasBSSSilicone oil 8 (36.4%) 1 (20%) 3 (14.3%) 0.85 (~20/140) 0.7 (~20/100) 0.53 (~20/70) 5 (22.7%)tamponade 1 (20%) 7 (33.3%) 1 and the 0 3 other 0.7 (~20/100) with 0.63 (~20/80)BSS; 3 (13.6%) the 5 (23.8%) hypotony 1 1 (20%) 0 1.18 (~20/300) 0

Eye 25-G pars plana vitrectomy for diabetic TRD MS Dikopf et al 1218

resolved within 1 week without further intervention. It was not seen in this series (possibly because of a small has been reported that 0.5-mm 25-G sclerotomy wounds number of patients in this group), they still realized heal more rapidly than 20-G wounds,16,24 perhaps significant improvements in central visual acuity. We supporting our observation that post 25-G PPV hypotony believe this to be clinically significant. 12,16 can self-resolve within a few days to a week. While strengths of this study include a large population Nonetheless, hypotony is a serious concern and must be of eyes with diabetic TRD operated on by a single surgeon closely monitored as it can lead to choroidal detachment, at a single institution, there are limitations. Postoperative 12 escape of gas, and inadequate tamponade, or increased acuity results were confounded by many variables— 22 susceptibility to infection. To prophylactically prevent differences in the amount of preoperative vitreous hypotony or other sequelae of would-be sutureless hemorrhage, chronicity of the RD, degree of PDR and wounds, all sclerotomies in our series were sewn. Of note, retinal vascular disease, worsening of preexisting those who prefer to leave sclerotomies sutureless have glaucoma, and development or removal of a cataract. advocated hyoptony-preventative techniques such as Conclusions of BCVA outcomes per type of tamponade 16,25 oblique sclerotomy incisions, conjunctival may also be limited by patient selection bias secondary fl displacement prior to cannula insertion, or uid-air/gas to location and severity of retinal detachment. Most 16 exchange at the completion of the case. importantly, however, the focus of this study on 25-G Increased risk of endophthalmitis with 25-G surgery PPV does not allow for a direct comparison with other has been also been noted, presumably from direct systems. Still, the results of this study suggest that 25-G inoculation of conjunctival bacteria during trocar PPV with SF6, C3F8, silicone oil, or BSS is a safe and insertion, decreased infusion rates during vitrectomy, effective modality for the treatment of complicated 16,22 and sutureless sclerotomies. In a large series, diabetic TRD. comparing endophthalmitis rates from 20- and 25-G vitrectomy surgeries, Kunimoto and Kaiser26 found a statistically significant increase in infection rates (0.018% vs 0.23%) in those undergoing 25-G vitrectomy. Summary However, many other large series have failed to find a What was known before – such a difference.27 29 It is still imperative to undertake K Twenty-five-gauge pars plana vitrectomy offers fi — measures to avoid this critical complication by either many bene ts over larger gauge systems namely tunnelling or angling the entry with the trocar or by reduced operative times, reduced post-operative inflammation, faster visual recovery, increased suturing the sclerotomy site at the end of surgery. In our patient comfort, reduced rate of intraoperative retinal series, a sclerotomies were sutured. No eyes developed break formation, avoidance of conjunctival dissection and endophthalmitis in this study, and other complications limbal stem cell damage, and less induced corneal with 25-G PPV in this study were relatively rare astigmatism. (Table 3). K Several small studies have investigated the use of 25-G vitrectomy for complicated retinal detachment, including Given the aforementioned benefits of 25-G PPV, diabetic tractional retinal detachment (TRD); comparable studies have evaluated its use in diabetic TRD repair. outcomes and complication rates between 25-G PPV (with Most studies were limited in size, retrospective, and silicone oil or octofluoropropane only) and 20-G or 23-G utilized primarily silicone oil.6,7,17–19 However, a systems have been found. prospective, randomized study by Kumar and colleagues compared the visual and anatomic success What this study adds K In our large series, 25-gauge vitrectomy was effective and rates of 23-G versus 25-G PPV with C3F8 or silicone oil safe in the repair of diabetic tractional retinal detachment, tamponade for repair of diabetic TRD (25 eyes in both including surgically complicated cases of combined groups). Complication rates were similar, and anatomic tractional and rhegmatogenous detachments or success was achieved in all 50 eyes in the study.20 Our proliferative vitreoretinopathy. study also demonstrates favorable outcomes of 25-G K Sulfur hexafluoride, octofluoropropane, silicone oil, or — balanced salt solution may all be safely and effectively PPV, yet with a wider range of tamponades SF6, used as tamponade agents in 25-gauge repair of diabetic C3F8, silicone oil, or BSS. Many eyes in our series had TRD. macula off detachments, combined rhegmatogenous detachments, or PVR, yet anatomic success was high, BCVA greatly improved for most patients, and Conflict of interest complication rates were low. The few eyes that were tamponaded with silicone oil (5 out of 70) in this study JIL has commercial relationship with QLT (F), Genentech had extensive and complicated detachments, and even (R), and Regeneron (R). The remaining authors declare no though statistical significance for postoperative BCVA conflict of interest.

Eye 25-G pars plana vitrectomy for diabetic TRD MS Dikopf et al 1219

References 15 Byeon SH, Chu YK, Lee SC, Koh HJ, Kim SS, Kwon OW. Problems associated with the 25-gauge transconjunctival 1 Rice TA, Michels RG, Rice EF. Vitrectomy for diabetic sutureless vitrectomy system during and after surgery. traction retinal detachment involving the macula. Am J Ophthalmologica 2006; 220: 259–265. Ophthalmol 1983; 95:22–33. 16 Chen E. 25-Gauge transconjunctival sutureless vitrectomy. 2 Fong DS, Ferris 3rd FL , Davis MD, Chew EY. Causes of Curr Opin Ophthalmol 2007; 18: 188–193. severe visual loss in the early treatment diabetic retinopathy 17 Shah CP, Ho AC, Regillo CD, Fineman MS, Vander JF, study: ETDRS report no. 24. Early Treatment Diabetic Brown GC. Short-term outcomes of 25-gauge vitrectomy Retinopathy Study Research Group. Am J Ophthalmol 1999; with silicone oil for repair of complicated retinal detachment. 127: 137–141. Retina 2008; 28: 723–728. 3 Regler R, Sachs HG, Hillenkamp J, Helbig H, Framme C. 18 Altan T, Acar N, Kapran Z, Unver YB, Ozdogan S. Long-term evaluation of anatomic and functional results Transconjunctival 25-gauge sutureless vitrectomy and after complicated retinal detachment treated with pars plana silicone oil injection in diabetic tractional retinal detachment. vitrectomy and heavy silicone oil tamponade. Klin Monbl Retina 2008; 28: 1201–1206. Augenheilkd 2009; 226: 707–712. 19 Park KH, Woo SJ, Hwang JM, Kim JH, Yu YS, Chung H. 4 Kirchhof B. Strategies to influence PVR development. Graefes Short-term outcome of bimanual 23-gauge transconjunctival Arch Clin Exp Ophthalmol 2004; 242: 699–703. sutureless vitrectomy for patients with complicated 5 Fujii GY, de Juan Jr E, Humayun MS, Pieramici DJ, Chang TS, vitreoretinopathies. Ophthalmic Surg Lasers Imaging 2010; 41: Awh C et al. A new 25-gauge instrument system for 207–214. transconjunctival sutureless vitrectomy surgery. 20 Kumar A, Duraipandi K, Gogia V, Sehra SV, Gupta S, Ophthalmology 2002; 109:1807–1812. Midha N. Comparative evaluation of 23- and 25-gauge 6 Lakhanpal RR, Humayun MS, de Juan Jr E, Lim JI, Chong LP, microincision vitrectomy surgery in management of diabetic Chang TS et al. Outcomes of 140 consecutive cases of 25-gauge macular traction retinal detachment. Eur J Ophthalmol 2014; transconjunctival surgery for posterior segment disease. 24: 107–113. Ophthalmology 2005; 112:817–824. 21 Holladay JT. Visual acuity measurements. J Cataract Refract 7 Sato T, Emi K, Bando H, Ikeda T. Faster recovery after 25 Surg 2004; 30: 287–290. gauge microincisionvitrectomy surgery than after 20-gauge 22 Bamonte G, Mura M, Stevie Tan H. Hypotony After vitrectomy in patients with proliferative diabetic 25-Gauge Vitrectomy. Am J Ophthalmol 2011; 151: 156–160. retinopathy. Clin Ophthalmol 2012; 6: 1925–1930. 23 Gupta OP, Weichel ED, Regillo CD, Fineman MS, Kaiser RS, 8 Narayanan R, Sinha A, Reddy RK, Krishnaiah S, Ho AC et al. Postoperative complications associated with Kupperman BD. Faster visual recovery after 23-gauge 25-guage pars plana vitrectomy. Ophthalmic Surg Lasers vitrectomy compared with 20-gauge vitrectomy. Retina 2010; Imaging 2007; 38: 270–275. 30: 1511–1514. 24 Keshavamurthy R, Ventkatesh P, Garg S. Ultrasound 9 Rizzo S, Genovesi-Ebert F, Murri S, Belting C, Vento A, biomicroscopy findings of 25 G transconjunctival sutureless Cresti F et al. 25-gauge, sutureless vitrectomy and standard (TSV) and conventional (20G) pars plana sclerotomy in the 20-gauge pars plana vitrectomy in idiopathic epiretinal same patient. BMC Ophthalmol 2006; 6:7. membrane surgery: a comparative pilot study. Graefes Arch 25 Hsu J, Chen E, Gupta O, Fineman MS, Garg SJ, Regillo CD. Clin Exp Ophthalmol 2006; 244: 472–479. Hypotony after 25-gauge vitrectomy using oblique versus 10 Neuhann IM, Hilgers RD, Bartz-Schmidt KU. Intraoperative direct cannula insertions in fluid-filled eyes. Retina 2008; 28: retinal break formation in 23-/25-gauge vitrectomy versus 937–940. 20-gauge vitrectomy. Ophthalmologica 2013; 229:50–53. 26 Kunimoto DY, Kaiser RS. Incidence of endophthalmitis after 11 de Paula AC, Nassaralla Junior JJ, Rodrigues FW, da Silva RE, 20- and 25-gauge vitrectomy. Ophthalmology 2007; 114: Correa e Silva RS, Barbosa Junior DR. Alterations in the 2133–2137. corneal topography after pars plana vitrectomy with the 27 Hu AY, Bourges JL, Shah SP, Gupta A, Gonzales CR, Oliver standard system of 20-gauge and the 25-gauge SC et al. Endophthalmitis after pars plana vitrectomy: a transconjunctival sutureless vitrectomy. Semin Ophthalmol 20- and 25-gauge comparison. Ophthalmology 2009; 116: 2012; 27:11–14. 1360–1365. 12 Thompson JT. Advantages and Limitations of Small Gauge 28 Scott IU, Flynn Jr HW, Dev S, Shaikh S, Mittra RA, Arevalo Vitrectomy. Surv Ophthalmol 2011; 56: 162–172. JF et al. Endophthalmitis after 25-gauge and 20-gauge pars 13 Ibarra MS, Hermel M, Prenner JL, Hassan TS. Longer-term plana vitrectomy: incidence and outcomes. Retina 2008; 28: outcomes of transconjunctival sutureless 25-gauge 138–142. vitrectomy. Am J Ophthalmol 2005; 139: 831–836. 29 Mason 3rd JO, Yunker JJ, Vail RS, White Jr MF, Feist RM, 14 Inoue M, Noda K, Ishida S, Nagai N, Imamura Y, Oguchi Y. Thomley ML et al. Incidence of endophthalmitis following Intraoperative breakage of a 25-gauge vitreous cutter. Am J 20-gauge and 25-gauge vitrectomy. Retina 2008; 28: Ophthalmol 2004; 138: 867–869. 1352–1354.

Eye